Agriculture Reference
In-Depth Information
varieties for the sake of giving isolates names that are familiar to international
researchers, than with investigating the threat that the pathogen may pose to local
crops. The latter is a much more important goal and should be emphasised. When a
pathogen is studied by several research groups, different authors may use a variety
of codes, either because they have used different differential sets or because they
wish to emphasise different features of their data. So long as a paper explains the
race codes clearly, there is no problem with several systems existing side-by-side.
Standardisation at the expense of scientific insight and practical application is not
desirable.
(c) Bulk isolates
A bulk isolate can be used to obtain approximate estimates of a virulence frequency
in the population from which it was sampled, by inoculating the isolate on a
differentially resistant host and a near-isogenic susceptible variety. The number of
successful infections on the former, divided by the number on the latter, is an
estimate of the frequency of virulent isolates in the bulk. It is not possible to
estimate associations between virulences in individual bulk isolates, but Wolfe
et al.
(1983) showed how bulk isolates could be used to estimate associations between
virulences in a population.
3.5.2 Responses to fungicides
(a) Frequencies of resistance
How the responses of a pathogen population to a fungicide should be summarised
depends on the nature of resistance to the chemical, the kind of samples tested, the
quantity estimated (i.e. ED
50
or MIC) and the use to which the data will be put. We
can distinguish three kinds of distribution of responses to fungicides: the 'all-
or-nothing' type, a continuous distribution of responses and the intermediate case of
several distinct levels of resistance. These three cases are all conveniently illustrated
by
B. graminis
: the all-or-nothing type by responses to QoIs (Sierotzki
et al.
, 2000a;
Chin
et al.
, 2001; Robinson
et al.
, 2002), the continuous distribution by responses of
B. graminis
f.sp.
tritici
to triadimenol (Robinson
et al.
, 2002; Wyand and Brown,
2005) and the several levels by the resistance of
B. graminis
f.sp.
hordei
to triazoles
and to fenpropimorph (Brown and Wolfe, 1990; Brown
et al.
, 1991a, 1991b, 1992,
1996; Brown and Evans, 1992; Brown, 1994a, 1996a; Blatter
et al.
, 1998; Wyand
and Brown, 2005). The response of
B. graminis
f.sp.
hordei
to ethirimol is either
sensitive or partially resistant (Brown and Wolfe, 1990; Brown
et al.
, 1991a) and is
controlled by a single gene (Brown
et al.
, 1992) but the quantitative difference
between the two responses is much smaller than that for the QoIs.
Samples may be single genotypes or bulk isolates, and the purpose of the tests
may either be to characterise the population itself or to develop advisory
recommendations. If single genotype isolates are tested, the severity of resistance of
an all-or-nothing kind is easily estimated as the frequency of resistant isolates, as for
